This is the mail archive of the binutils@sources.redhat.com mailing list for the binutils project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [patch] MIPS: Fix synthesized doubleword transfers (ping)

From: "Maciej W. Rozycki" <macro at linux-mips dot org>
To: cgd at broadcom dot com
Cc: rsandifo at redhat dot com, binutils at sources dot redhat dot com
Date: Fri, 25 Feb 2005 15:45:52 +0000 (GMT)
Subject: Re: [patch] MIPS: Fix synthesized doubleword transfers (ping)
References: <Pine.LNX.4.61L.0502242001360.23589@blysk.ds.pg.gda.pl><87sm3lhdo1.fsf@firetop.home> <mailpost.1109279149.22995@news-sj1-1><yov51xb539ng.fsf@xl-sj1-01.sj.broadcom.com>

On Thu, 24 Feb 2005 cgd@broadcom.com wrote:

> > I suppose bad things could happen if you composed an aligned
> > address from an unaligned symbol and an unaligned offset,
> > but is that allowed?
> 
> yes, at least as far as the ISA is concerned.

 For any ISA actually -- that MIPS IV limitation on address bases and 
displacements was essentially a hoax -- AFAIK, no MIPS IV processor has 
implemented it (including the R8000) and it's been explicitly removed for 
the MIPS64 ISA.  Besides, it would have broken binary compatibility with 
MIPS II and MIPS III anyway.

 It was probably meant as a way to allow some address calculation 
optimisation, which has never actually happened.

  Maciej

References:
- [patch] MIPS: Fix synthesized doubleword transfers (ping)
  - From: Maciej W. Rozycki
- Re: [patch] MIPS: Fix synthesized doubleword transfers (ping)
  - From: Richard Sandiford
- Re: [patch] MIPS: Fix synthesized doubleword transfers (ping)
  - From: cgd

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]